Telephone: 81-744-29-8863; FAX: 81-744-24-6866
Plasma ADAMTS13 activity may predict early adverse events in living donor liver transplantation: Observations in 3 cases
Article first published online: 9 MAR 2006
Copyright © 2006 American Association for the Study of Liver Diseases
Volume 12, Issue 5, pages 859–869, May 2006
How to Cite
Ko, S., Okano, E., Kanehiro, H., Matsumoto, M., Ishizashi, H., Uemura, M., Fujimura, Y., Tanaka, K. and Nakajima, Y. (2006), Plasma ADAMTS13 activity may predict early adverse events in living donor liver transplantation: Observations in 3 cases. Liver Transpl, 12: 859–869. doi: 10.1002/lt.20733
- Issue published online: 20 APR 2006
- Article first published online: 9 MAR 2006
- Manuscript Accepted: 7 JAN 2006
- Manuscript Received: 4 OCT 2005
- Japanese Ministry of Education, Culture, and Science
- Ministry of Health and Welfare of Japan for Blood Coagulation Abnormalities. Grant Number: H17-02
A disintegrin-like and metalloproteinase with thrombospondin type-1 motifs 13 (ADAMTS13) is a metalloproteinase that specifically cleaves the multimeric von Willebrand factor (VWF). Deficiency of ADAMTS13 increases the unusually large VWF multimers (UL-VWFM), which leads to platelet clumping and/or thrombus formation, resulting in microcirculatory disturbance. We serially determined the activity of plasma ADAMTS13, together with VWF antigen (VWF:Ag) and UL-VWFM, in association with the development of early graft dysfunction in 3 liver transplant recipients and 4 patients with major hepatectomy as controls. In case 1, ADAMTS13 activity decreased markedly from 108% to less than 3% with concomitant thrombocytopenia on posttransplantation day 7, when acute rejection occurred. Simultaneously, UL-VWFM were detected. During the second episode of rejection, VWF:Ag increased to 368% with the appearance of UL-VWFM, while ADAMTS13 activity was as low as 18%, indicating an imbalance between a large amount of UL-VWFM and low activity of ADAMTS13. Administration of fresh frozen plasma (FFP) together with treatment for acute rejection resulted in an improvement of ADAMTS13 activity and disappearance of the UL-VWFM. In case 2, ADAMTS13 activity promptly decreased to 9% with thrombocytopenia on day 1, when ischemia-reperfusion injury occurred. Subsequently, the ADAMTS13 activity increased steadily without appearance of UL-VWFM, and the patient recovered uneventfully. ADAMTS13 activity decreased to 15% immediately after transplantation in case 3 as well. In contrast, ADAMTS13 activity never decreased below 20% in 4 patients with major hepatectomy as controls. In conclusion, these results indicate that the kinetics of ADAMTS13 and UL-VWFM could be good indicators of adverse events after liver transplantation. Our findings not only suggest a novel mechanism for thrombocytopenia, but also provide a useful tool for diagnosis of graft dysfunction in the early stage after transplantation. Liver Transpl 12:859–869, 2006. © 2006 AASLD.
Thrombotic microangiopathies (TMAs) are defined as life-threatening generalized disorders, characterized by microangiopathic hemolytic anemia, destructive thrombocytopenia, and organ dysfunction caused by microvascular platelet thrombi.1, 2 Because of these features, TMAs are usually expressed heterogeneously, and can include thrombotic thrombocytopenic purpura (TTP) with neurotropic signs prevalent in adults, but not exclusively, and hemolytic-uremic syndrome with predominant nephrotropic signs.1, 2 TMA is also recognized as a critical complication after solid organ transplantation.3–9 However, most cases of transplantation-associated TMA are not clearly distinguishable as either TTP or hemolytic-uremic syndrome.
The discovery of a disintegrin-like and metalloproteinase with thrombospondin type-1 motifs 13 (ADAMTS13) has provided a breakthrough in our understanding of TMA pathogenesis. Recent studies indicate that ADAMTS13 is produced mainly in the liver, exclusively the stellate cells (formerly called Ito cells), and then thought to be released into the circulation via the microsinusoidal system,10 where ADAMTS13 specifically cleaves multimeric von Willebrand factor (VWF) between Tyr1605 and Met1606 in the A2 domain.11–14 VWF is synthesized in vascular endothelial cells, and released into the plasma as unusually large VWF multimers (UL-VWFM), which have potent biological activities.2, 15 Under physiological conditions, UL-VWFM are rapidly degraded into smaller VWF multimers by ADAMTS13.2, 15 Deficiency of the protease increases the level of UL-VWFM in plasma and leads to platelet aggregation and/or thrombus formation, finally resulting in TTP.16, 17 In fact, the activity of ADAMTS13 is significantly decreased in most patients with TTP, whereas it is relatively preserved in the majority of patients with hemolytic-uremic syndrome.18
On the other hand, thrombocytopenia is commonly observed during the first week after liver transplantation, with or without apparent TMA.19–21 Some clinical studies have demonstrated a significantly poorer prognosis in recipients with severe thrombocytopenia than in those without,22, 23 suggesting a close relationship of thrombocytopenia to allograft dysfunction including ischemia-reperfusion injury and acute rejection, which are common adverse events in the early period after transplantation. The primary target for these adverse events is vascular endothelial cells, and injury to these cells in the graft liver results in a large amount of VWF production.24–27 It is known that circulating VWF levels are markedly high in recipients with poor early graft function.27 Platelet adhesion to the sinusoidal endothelium with a concomitant increase of VWF expression in the reperfused liver is one of the main deleterious effects of cold preservation of liver allografts.26
A few reports have described that ADAMTS13 activity can be used as a marker to diagnose TMA in recipients of liver transplants and renal allografts.3, 4 However, there is little information about the relationship between ADAMTS13 and allograft dysfunction and thrombocytopenia after liver transplantation. In the present study of 3 living-donor liver transplant recipients, we measured the plasma activity of ADAMTS13 together with VWF and UL-VWFM, and thereby attempted to clarify a potential role of the protease activity in adverse events including ischemia-reperfusion injury and/or acute rejection. As controls, 4 patients with major hepatectomy were also analyzed.
Plasma levels of ADAMTS13 activity, VWF antigen (VWF:Ag), and UL-VWFM were sequentially evaluated before and after liver transplantation in 3 recipients. Inhibitor activity against ADAMTS13 was assayed on day 7 after transplantation in case 1, and on day 1 after transplantation in case 2. Also, as controls, plasma ADAMTS13 activity and VWF:Ag were measured in 4 patients with normal livers who underwent major hepatectomy in our hospital during July to November 2005. Blood was taken in plastic tubes containing a 1/10th volume of 3.8% sodium citrate, and platelet-poor plasma prepared by centrifugation at 3,000g at 4°C for 15 minutes was stored in aliquots at −80°C until analysis. The activity of plasma ADAMTS13 was assayed by the following 2 methods: 1) VWF-multimer assay using intact VWFM as a substrate according to Furlan et al.,28 with slight modification.18 The detection limit of the activity with this method was 3%, and the level obtained for 60 normal subjects was 102 ± 23% (mean ± standard deviation).18 2) Novel enzyme-linked immunosorbent assay using a murine monoclonal antibody specifically recognizing Tyr1605 residue of VWF-A2 domain,29 generated by ADAMTS13 cleavage, and a recombinant GST-VWF73-His polypeptide30 as a substrate. The detection limit of the activity with this enzyme-linked immunosorbent assay was 0.5%, and the normal level obtained for 55 healthy individuals was 99.1 ± 21.5% (mean ± standard deviation).29 In the Case Reports and Results section and the Discussion section below, the values determined by VWF-multimer assay are described. Plasma UL-VWFM was evaluated by vertical agarose gel electrophoresis according to the method of Warren et al.,31 with modifications. The activity of inhibitor against ADAMTS13 was evaluated using heat-inactivated plasma at 56°C for 30 minutes.16, 17 Plasma VWF:Ag was measured by a sandwich enzyme immunoassay using a rabbit anti-human VWF polyclonal antibody. The value obtained for healthy subjects in our laboratory (n = 54; 30 males, 24 females, 20-39 yr of age) was 100 ± 53% (mean ± standard deviation).
CASE REPORTS AND RESULTS
A 27-yr-old man with a diagnosis of Budd-Chiari syndrome was referred to our department for liver transplantation on October 5, 2004. Radiological imaging showed complete occlusion of both the middle and left hepatic veins, together with stenosis of the orifice of the right hepatic vein. The liver had rapidly swollen because of severe liver congestion, and massive ascites was noted before transplantation. Preoperative blood examination revealed a serum total bilirubin level of 4.8 mg/dL, hemoglobin of 10.9 gm/dL, platelet count of 83,000/μL, creatinine of 0.7 mg/dL, and alanine aminotransferase (ALT) activity of 23 IU/L (Fig. 1). His prothrombin time was 18 seconds (international normalized ratio: 1.49) and bleeding time was 5 minute. Anticoagulation factors protein C (54%), protein S (56%), and antithrombin III (75%) were not severely decreased. He had no history of hematological disorders, thrombotic events or relevant family diseases. On November 22, the patient underwent living donor liver transplantation, receiving a cross-match-negative and blood group type-identical right liver graft from his brother, who was healthy and had no history of previous hematological disorders or relevant family diseases. The graft weight and graft-to-recipient weight ratio were 588 gm and 0.93%, respectively. Operative blood loss was 8,410 mL and required 12 units of packed red blood cells and 10 units of fresh frozen plasma (FFP). Platelet concentrate was not required. Posttransplantation immunosuppressive treatment consisted of tacrolimus and methylprednisolone. The dose of tacrolimus was adjusted to maintain whole-blood trough levels of 10-15 ng/mL.
Before transplantation, the activity of ADAMTS13 was 108% of the normal control activity (Figs. 1 and 2), VWF:Ag was 175% (Fig. 1), and UL-VWFM was not detected (Fig. 3). During an uneventful course in the early stage after transplantation, the platelet count decreased gradually to 62,000/μL on postoperative day 5, and reached a nadir (25,000/μL) on day 7, when ADAMTS13 activity decreased markedly to less than 3% (Fig. 2), although no inhibitor against the protease was detected. Simultaneously, the serum ALT level increased to 124 IU/L on day 1 because of ischemia-reperfusion injury, decreased thereafter to 97 IU/l on day 5, and again increased rapidly to 175 IU/L on day 6 due to acute rejection, which was clinically diagnosed (Fig. 1). VWF:Ag was mildly increased to 188% on day 7. UL-VWFM was detectable even on day 1, diminished gradually during days 2 to 4, and again became evident on day 7 (Fig. 3). Based on the activity of ADAMTS13, we considered that TMA was the cause of thrombocytopenia, but the recipient never showed any apparent clinical features including renal dysfunction, neuropsychological symptoms or hemolytic anemia. We therefore administered a large amount of FFP (4 to 20 units daily) from day 7 to day 30 to restore the ADAMTS13 activity. Plasmapheresis was not performed, because no inhibitor against ADAMTS13 was detected. Bolus injection of methylprednisolone (500 mg daily) was added to treat the acute rejection from day 7 to 9, without conversion of tacrolimus to another drug. Thereafter, the platelet count increased gradually to 90,000/μL on day 20 without administration of platelet concentrate. The activity of ADAMTS13 also increased to 12% on day 14 (Fig. 2), and this was maintained until day 65. After the first episode of acute rejection around day 7, VWF:Ag increased further and reached 368% on day 21, when ALT level again increased to 259 IU/L due to a second episode of acute rejection. The amount of UL-VWFM diminished transiently on day 9 during remission of the first acute rejection episode, but increased again on day 11, coinciding with a mild elevation of transaminase. After the amount of UL-VWFM diminished on day 15, it became prominent again on day 22 during the second episode of acute rejection (clinically diagnosed). Bolus injection of methylprednisolone quickly reduced the level of ALT, and the platelet count gradually increased and reached 199,000/μL on day 30 (Fig. 1). ADAMTS13 increased gradually to 50%, corresponding to the lower limit of the normal range, and VWF:Ag decreased gradually and returned to the normal range at the time of discharge on day 98 (Fig. 1). UL-VWFM became undetectable until day 45 (Fig. 3). After liver transplantation, the patient lost a large amount of lymphatic fluid via the peritoneal drain, but this gradually decreased and had disappeared completely by about day 65.
A 53-yr-old man with a diagnosis of hypercitrullinemia type II was referred to our department for liver transplantation on January 4, 2005. The activity of argininosuccinate synthetase was markedly low (0.36 U/gm liver; normal range 2.59 ± 1.13 U/gm liver), while the quantity of the enzyme was preserved (0.0052 U/mg protein; normal range 0.0033 ± 0.0012 U/mg protein). The serum ammonia level was sometimes higher than 600 μg/dL, and analysis of plasma amino acid showed a markedly high concentration of citrulline (338 nmol/mol, normal range 17-43 nmol/mol). The patient had several episodes of deep hepatic coma. On February 14, 2005, he underwent auxiliary partial orthotopic liver transplantation using a left liver graft from his wife, because his own liver function was normal except for amino acid metabolism. The day 0 biopsy of the donor liver revealed macrovesicular steatosis. The graft weight and graft-to-recipient weight ratio were 304 gm and 0.56%, respectively. Operative blood loss was 2,900 mL and required 8 units of packed red blood cells and 4 units of FFP. Platelet concentrate was not required. Posttransplantation immunosuppressive treatment consisted of tacrolimus and methylprednisolone, as used in case 1.
Preoperative blood examination revealed a serum total bilirubin level of 1.9 mg/dL, hemoglobin 13.8 gm/dL, platelet count 142,000/μL, creatinin 0.7 mg/dL, and ALT 85 IU/L (Fig. 4). The coagulation parameters were normal before transplantation, and had no history of hematological disorders or relevant family diseases. Pre-transplant ADAMTS13 activity was 110% of the normal control activity, VWF:Ag was 142%, and UL-VWFM was undetectable (Fig. 5). On day 1 after transplantation, ALT increased markedly to 1226 IU/L probably because of ischemia-reperfusion injury (Fig. 4). The platelet count decreased to 51,000/μL on postoperative day 1 and reached a nadir of 38,000/μL on day 3. The activity of ADAMTS13 decreased markedly to 9% on day 1, but no inhibitor against ADAMTS13 was detected. VWF:Ag decreased slightly to 89%. UL-VWFM was not detected (Fig. 5). Although the patient had no signs of TMA including renal dysfunction, neuropsychological symptoms, or hemolytic anemia, a small dose of FFP was administered from days 1 to 11 after transplantation to restore the activity of ADAMTS13. Thereafter, ALT promptly decreased, and the platelet count increased gradually to 182,000/μL on day 14. The levels of blood ammonia and citrulline were normalized within 2 weeks after transplantation, suggesting that the auxiliary partial liver graft was functioning very well. The activity of ADAMTS13 increased gradually to 31% on day 7, 42% on day 14, and reached 102% on day 44. VWF:Ag did not increase markedly, and UL-VWFM was never detected during hospitalization (Fig. 5).
A 57-yr-old man with a diagnosis of hepatitis B virus-related cirrhosis was referred to our department for liver transplantation on June 6, 2005. He had massive uncontrollable ascites and marked jaundice. Preoperative blood examination revealed a serum total bilirubin level of 4.8 mg/dL, hemoglobin of 10.9 gm/dL, platelet count of 83,000/μL, creatinine of 0.7 mg/dL, and ALT of 23 IU/L (Fig. 1). His prothrombin time was 22.6 seconds (international normalized ratio: 1.97) and his Model for End-Stage Liver Disease score was 25. He had no history of hematological disorders or relevant family diseases. On July 6, the patient underwent living donor liver transplantation, receiving a cross-match-negative and blood group type-compatible (from B donor to AB recipient) right liver graft from his son, who was healthy and had no history of previous hematological disorders or relevant family diseases. The graft weight and graft-to-recipient weight ratio were 936 gm and 1.43%, respectively. Operative blood loss was 24,750 mL and required 77 units of packed red blood cells, 70 units of FFP and 5 units of platelet concentrate. The initial posttransplantation immunosuppressive treatment consisted of tacrolimus and methylprednisolone. The dose of tacrolimus was adjusted to maintain whole-blood trough levels of 10-15 ng/mL. Tacrolimus was converted to cyclosporine on day 12 due to tacrolimus-induced leukoencephalopathy, which was diagnosed by magnetic resonance imaging, and recovered completely after conversion to cyclosporine. The dose of cyclosporine was adjusted to maintain whole-blood trough levels of 150-200 ng/mL. The patient was positive for the YMDD-mutant of hepatitis B virus before transplantation and treated with lamivudine and adefovir before and after transplantation. Prophylactic infusion of human hepatitis B immunoglobulin was administered intravenously to prevent hepatitis B recurrence according to the reported protocol.32 Hepatitis B did not recur and the hepatitis B virus-deoxyribonucleic acid levels were below the detection limit after transplantation during observation.
Before transplantation, the activity of ADAMTS13 showed a low level (32%) probably because of severely impaired hepatic functional reserve (Fig. 6). VWF:Ag was markedly increased to more than 500% (Fig. 6), but UL-VWFM was not detected before transplantation (Fig. 7). While initial elevation of ALT due to ischemia-reperfusion was minimal, ADAMTS13 activity decreased markedly to 14% on day 2 after transplantation. His platelet count decreased to 13,000/ μL during the first week, but platelet concentrate was never administered. VWF:Ag level significantly decreased on day 1 after transplantation, and increased gradually until day 6. ADAMTS13 activity begun to increase from day 3, but again markedly decreased to 8% on day 13 when hemolysis due to B cell-mediated graft-vs.-host disease occurred. This hemolysis was considered to be caused by donor-derived antibody against anti-blood type A antigen, because anti-blood type A antigen appeared on day 13 in the peripheral blood. Rapid decrease of hemoglobin associated with increase of total bilirubin and lactate dehydrogenase was observed, but ALT remained normal. Serum creatinine level was not elevated significantly. During this episode, increase of VWF:Ag to 335% was seen, but UL-VWFM was not up-regulated (Fig. 7). This hemolytic reaction was successfully treated with bolus injection of steroid. ADAMTS13 increased to 25% until day 21. Decrease of VWF:Ag was also seen after this treatment. ALT increased mildly around day 25, but we could not determine the specific reason. This increase of ALT recovered without treatment. ADAMTS13 did not increase during this episode, and reached 48% on day 90.
Patients With Major Hepatectomy
Four patients who underwent major hepatectomy were analyzed for ADAMTS13 activity and for VWF:Ag before and after their operations, for comparison to the liver transplant patients. These 4 patients had normal hepatic parenchyma without cirrhotic change. While VWF:Ag levels were rather higher in these patients than those in cases 1-3 with liver transplantation, ADAMTS13 never decreased below 20% even in the very early phase after operation (Fig. 8).
In the present study, we serially determined the plasma values of ADAMTS13 activity, VWF:Ag, and UL-VWFM, and demonstrated their relationship to early adverse events including ischemia-reperfusion injury and/or acute graft rejection after liver transplantation. In case 1, the activity of plasma ADAMTS13 decreased markedly to less than 3% with concomitant thrombocytopenia on day 7 after transplantation, when acute rejection occurred. UL-VWFM was detected at the time of ischemia-reperfusion injury and also during 2 episodes of acute rejection. In case 2, the activity of the protease decreased markedly to 9% with concomitant thrombocytopenia on day 1, when ischemia-reperfusion injury was observed. The immediate marked decrease of ADAMTS13 was observed also in case 3. On the other hand, changes of ADAMTS13 levels were milder in patients with major hepatectomy than in liver transplant patients. These results indicate that decreased ADAMTS13 activity and the appearance of UL-VWFM are closely related to the development of early posttransplantation allograft dysfunction accompanied by thrombocytopenia.
We analyzed 3 liver transplant recipients, and all of these patients showed a significant decrease of ADAMTS13 with or without upregulation of UL-VWFM during adverse events after transplantation. However, no patient showed the typical clinical signs of TMA including neurological disorder or renal dysfunction during the significant drop in ADAMTS13, while various degrees of thrombocytopenia were associated with the drop in ADAMTS13. These results demonstrate that decrease of ADAMTS13 often occurs after liver transplantation without apparent clinical signs of TMA, and this phenomenon may have a functional relevance to the deterioration of the liver graft presumably due to the mechanism of local TMA within the graft site.
Posttransplantation thrombocytopenia is more often associated with early acute rejection in liver transplant recipients.22 As to the mechanism of the thrombocytopenia, several factors involving sequestration of platelets in the reperfused liver graft, immunologic reactions, increased platelet consumption, reduced platelet production, impaired production of thrombopoietin, medication, or a combination of these have been suspected,19–22, 33, 34 but our findings appear to suggest a novel mechanism of not only early posttransplantation thrombocytopenia, but also early graft dysfunction, which exerts a great influence on the prognosis of recipients.22–25, 27
In case 1, we were unable to evaluate the activity of ADAMTS13 during the first few days after surgery, but it could have been low at the time of ischemia-reperfusion injury because of the apparent presence of UL-VWFM on day 1. In case 2, the marked decrease in ADAMTS13 activity on day 1 may have been due to consumption of the protease because of the weaker VWFM and lower plasma level of VWF compared with the situation before surgery. During the very early stage after transplantation (on day 1), it was noticeable that UL-VWFM was increased in case 1, but decreased in case 2. The decrease of UL-VWFM in case 2 may be explained by the consumption of UL-VWFM in the process of platelet aggregation, probably due to more severe ischemia-reperfusion injury than that in case 1.28 Similarly, the decrease of UL-VWF in plasma was observed in patients with TTP during relapse.35 Another possible explanation may have been the difference in the amount of UL-VWFM released from the damaged sinusoidal endothelium during the surgical procedures, including extirpation of the native diseased liver; case 1 with advanced cirrhosis may have produced a larger amount of UL-VWFM than case 2 with a normal liver, because sinusoidal endothelial damage with capillarization is further augmented as liver fibrosis progresses.36, 37 Regarding the mechanism responsible for the decrease in ADAMTS13 after liver transplantation, no antibody against ADAMTS13 was detected in either case. Therefore, we speculate that consumption of the protease due to a large amount of UL-VWFM, as seen in case 2, and unknown factors such as proinflammatory cytokines including interleukin-6, which might have inhibited the action of ADAMTS13, were involved.38 In fact, the plasma concentration of interleukin-6 was significantly increased and reached a peak 2 hours after reperfusion of the liver graft.39
In case 3, plasma VWF:Ag was very high before the operation, probably because of sinusoidal endothelial injury due to severe liver cirrhosis.36, 37 The VWF:Ag rapidly decreased on day 1 after transplantation and UL-VWFM was not detected. We speculate that this significant and immediate decrease of VWF:Ag would be due to the washout effect of a large amount of transfusion during operation. That might be the reason why the VWF:Ag increased gradually during the first postoperative week in spite of the uneventful early recovery of the patient. Although transfusion of a large amount of FFP (70 units) during the operation in case 3 might have increased ADAMTS13 activity, the ADAMTS13 activity on day 1 decreased significantly to 15%. This result suggest that consumption of ADAMTS13 due to ischemia-reperfusion injury during liver transplantation would be so much as to wipe out the effect of 70 units of FFP.
The result of case 2 may be quite important in this study, because the native normal liver was preserved in this patient. The initial drop of ADAMTS13 activity immediately after transplantation was also significant in case 2 as well as case 3, although the production of ADAMTS13 might be preserved by the native right liver with auxiliary partial orthotopic liver transplantation in case 2. This result may suggest that consumption of the ADAMTS13 plays more important role in decrease of ADAMTS13 activity after transplantation during adverse events than decreased ADAMTS13 production due to impaired liver allograft function. Therefore, ADAMTS13 may decrease in the initial phase of adverse events before impairment of graft function.
Interestingly, in case 1, VWF:Ag was markedly increased to 368% and UL-VWFM was detected during the second episode of rejection on day 21, while ADAMTS13 activity was as low as 18% and the platelet count was relatively well maintained at 75,000/μL. These findings indicate that the imbalance of a large amount of UL-VWFM relative to low ADAMTS13 activity could be a good indicator of allograft rejection, even in the absence of severe thrombocytopenia. Thus, our results may be able to explain the fact that liver transplant recipients with increased levels of circulating VWF, a reliable marker of endothelial damage,40, 41 show poor early graft function.17
In case 3, ADAMTS13 decreased during hemolysis by B cell-mediated graft-vs.-host disease. However, UL-VWFM was not upregulated during this episode. Differing from ischemia-reperfusion injury or acute rejection, this hemolytic reaction was a systemic reaction due to antibody against blood type A antigen. ADAMTS13, VWF:Ag, and UL-VWFM did not change during nonspecific elevation of ALT around day 25 in case 3. It may be important to analyze ADAMTS13 in combination with UL-VWFM to detect liver transplantation-specific adverse events.
The decrease of ADAMTS13 was milder in patients with major hepatectomy in comparison to liver transplant patients, while VWF:Ag increased higher. This result suggests that significant decrease of ADAMTS13 below 20% would be a liver transplantation-specific event, while the mechanism of difference in changes of ADAMTS13 activity between hepatectomy and liver transplantation is unknown.
The primary target of ischemia-reperfusion injury and allograft rejection is the sinusoidal endothelial cells of the liver graft.24–26, 42 Deposition of activated platelets on the sinusoidal endothelium with a concomitant increase of VWF expression have been found in the liver immediately after reperfusion or cold preservation.24, 25 In addition, upregulated VWF expression has been observed in liver allografts during acute rejection.25 Furthermore, recipients with acute rejection show enhanced cytokinemia including tumor necrosis factor-α, which leads to endothelial activation and stimulates the release of UL-VWFM from endothelial cells.38, 42 VWF, the substrate of ADAMTS13, synthesized in vascular endothelial cells, mediates the initiation and progression of thrombus formation at sites of vascular injury.14, 40, 41, 43 VWF is released into plasma as UL-VWFM, which has high platelet aggregation activity. The deficiency of ADAMTS13, together with the excessive release of UL-VWFM from injured graft endothelial cells observed in our study, may cause sinusoidal microcirculatory disturbance and subsequent graft dysfunction.
Various degrees of thrombocytopenia were commonly observed after liver transplantation, especially during the first postoperative week, and many of these patients recover without specific treatment. However, it might be possible that thrombocytopenia is a sign of deterioration of the liver graft in some of the patients with thrombocytopenia, because clinical studies demonstrated that thrombocytopenia was associated with poor prognosis.22–25, 27 If thrombocytopenia is combined with significant decrease of ADMTS13, liver graft function may be deteriorated via the TTP like mechanism due to microcirculatory disturbance. Therefore, monitoring of ADAMTS13 would be quite important to judge the necessity of treatment for thrombocytopenia. In case 1, we administered a large amount of FFP from day 7 to prevent further deterioration of thrombocytopenia with TMA mechanism. In cases 2 and 3, a limited dose of FFP was administered as a prophylaxis of graft dysfunction due to TMA-like reaction, because the ADAMTS13 activity significantly decreased. However, it is to be elucidated whether such prophylactic use of FFP based on the ADAMTS13 activity would provide a beneficial effect.
The values of decreased plasma ADAMTS13 activity by VWF-multimer assay in the present study do not appear to be influenced by the elevated plasma UL-VWFM, because the comparable results are drawn by the novel enzyme-linked immunosorbent assay, which is totally insensitive to the presence of intact VWFM.29 This new enzyme-linked immunosorbent assay method would be very useful in clinical application of ADAMTS13 monitoring in liver transplantation, because the results can be obtained within several hours.
At present, FFP is the only available source of ADAMTS13 replacement.18 Remarkably, in our patients, infusion of FFP, but not platelet concentrate, resulted in gradual improvement of severe thrombocytopenia together with an increase in ADAMTS13 activity. This is an extremely important issue in the treatment of thrombocytopenia associated with allograft dysfunction after liver transplantation, because administration of platelet concentrate under pathological conditions, including an imbalance between decreased ADAMTS13 activity and enhanced VWF production, would further exacerbate the formation of platelet aggregates mediated by uncleaved UL-VWFM, leading to multiorgan failure, as seen in TMA.2 Platelet concentrate was never administered in case 3 even when the platelet count decreased to 13,000/μL on days 4 to 6, because the activity of ADAMTS13 was low. The mechanism of thrombocytopenia associated with early adverse events after liver transplantation is noteworthy. In the posttransplantation period, patients are especially susceptible to TMA because of administration of calcineurin inhibitors including tacrolimus and cyclosporine, which are well-documented to induce TMA.44 In case 1, we successfully treated the thrombocytopenia by administering high-dose FFP without conversion of calcineurin inhibitors. Therefore, it would be particularly useful to determine the values of ADAMTS13 and its substrate, VWF:Ag, together with UL-VWFM in the early period after transplantation, not only for the diagnosis of TMA, but also for clarifying the mechanism of thrombocytopenia. Our experience, although based on only 3 liver transplantations and 4 major hepatectomy cases, may provide useful data that are relevant to the diagnosis and treatment of ischemia-reperfusion injury and allograft rejection, as well as for clarifying the pathogenesis of thrombocytopenia after liver transplantation.
- 1Acquired platelet disorders. In: BloomAL, ForbesCD, ThomasTP, TuddenhamEGD, eds. Hemostasis and Thrombosis. London: Churchill-Livingstone, 1994: 767–815., .
- 29Monoclonal antibodies to a VWF-A2 decapeptide with the C-terminal residue Tyr1605, generated by ADAMTS13 cleavage, develops a highly sensitive ELISA for its activity and characterize Upshaw-Schulman syndrome. Blood 2005; 106(Abs# 2643)., , , , .
- 37Studies on capillarization of the hepatic sinusoids in alcoholic liver disease. Alcohol Alcohol 1993; 28: 77–84., , , , .